A control method includes empirically constructing for a flexible manipulator an estimated mapping between an end-effector movement and a plurality of values representing movement of at least one actuator, measuring an actual end-effector movement and an actual movement of the at least one actuator, and, based on the measuring, updating the estimated mapping while the flexible manipulator is active such that the mapping is adapted to the present environment of the flexible manipulator.
Legal claims defining the scope of protection, as filed with the USPTO.
1. A control method, comprising: empirically constructing for a flexible manipulator an estimated mapping between a movement of at least one actuator and a movement of an end-effector of the flexible manipulator; measuring an actual movement of the at least one actuator and a resulting actual end-effector movement; and based on the measuring, updating the estimated mapping while the flexible manipulator is active such that the mapping is adapted to a present environment of the flexible manipulator.
2. The method of claim 1 , further comprising: identifying a desired end-effector movement; determining at least one actuator control signal value based on the estimated mapping and the desired end-effector movement; providing the at least one actuator control signal value; and after providing the at least one actuator control signal value, measuring the actual end-effector movement and the actual movement of the at least one actuator.
3. The method of claim 1 , the updating including using an optimization function.
4. The method of claim 3 , the optimization function minimizing a change in the estimated mapping due to the updating.
5. The method of claim 4 , the minimizing achieved by minimizing a norm of the mapping, wherein the norm is one of a Frobenius norm, an L1-norm, an L2-norm and a Huber norm.
6. The method of claim 3 , the optimization function maximizing a change in the estimated mapping due to the updating.
7. The method of claim 3 , including using at least one restraint on the optimization.
8. The method of claim 7 , the restraint being one of a weighting of at least one portion of the mapping, a restraint related to a geometry of the flexible manipulator, and a restraint related to a geometry of the at least one actuator.
9. The method of claim 1 , wherein the mapping is a linear mapping.
10. The method of claim 1 , wherein the mapping is a non-linear mapping.
11. The method of claim 1 , wherein the mapping is a quadratic mapping.
12. The method of claim 1 , wherein the mapping is a Jacobian mapping.
13. The method of claim 1 , wherein updating the estimated mapping while the flexible manipulator is active expands a reachable workspace of the manipulator.
14. The method of claim 1 , wherein updating the estimated mapping while the flexible manipulator is active avoids inverted mappings induced by the environment, thereby avoiding instability and positive feedback loops.
15. The method of claim 1 , wherein empirically constructing the estimated mapping includes, for each of the at least one actuator: providing values in a sequence to effectuate movement of the actuator; after each value in the sequence of values, measuring a resulting movement of the actuator and a resulting movement of the end-effector; and calculating a mapping relating expected actuator movement to expected end-effector movement based on the measured actuator movement and the measured end-effector movement.
16. A controller for a flexible manipulator, comprising: an input interface coupled to a plurality of sensors, the sensors configured to sense movement of at least one actuator and an end-effector of the flexible manipulator; an output interface; and a processor in communication with the input interface and the output interface, the processor configured to: provide at least one expected actuator movement value to the output interface; receive from the input interface information representing an actual actuator movement and a resulting actual end-effector movement; and update an estimated mapping between actuator movement and end-effector movement based on the received information; wherein the estimated mapping is updated while the flexible manipulator is active such that the flexible manipulator adapts to its present environment.
17. The controller of claim 16 , wherein the sensors include at least one position sensor.
18. The controller of claim 16 , wherein the sensors include at least one accelerometer.
19. The controller of claim 16 , wherein the mapping is a Jacobian mapping.
20. A non-transitory computer-readable medium comprising computer-executable instructions to: identify a desired end-effector movement for a flexible manipulator; determine from a predefined mapping at least one value for actuating at least one actuator, the value estimated to effectuate the desired end-effector movement by actuating the at least one actuator; receive information representing an actual movement of the at least one actuator and an actual movement of the end-effector resulting from application of the at least one value; and update the mapping based on the received information to adapt the flexible manipulator to its present environment.
21. The non-transitory computer-readable medium of claim 20 , further comprising computer-executable instructions to: provide a plurality of values in a sequence; receive information representing movement of the at least one actuator and movement of the end-effector in response to providing the sequence of values; construct a mapping relating actuator movement to end-effector movement; and store the mapping as the predefined mapping.
22. The non-transitory computer-readable medium of claim 20 , the computer-executable instructions to update the mapping including computer-executable instructions to use an optimization function with at least one restraint.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
March 5, 2014
November 8, 2016
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.